We are attempting to assess the efficacy of two behavioral methods to control alcoholism among methadone treatment patients. In Experiment 1, alcoholic methadone patients will receive methadone contingent upon their ingestion of disulfiram. In Experiment 2, a negative blood-alcohol reading will be required before the patient receives his daily methadone. In both experiments, the incidence of drinking and other maladaptive behavior will be recorded. The results of these two treatments will be compared with the data obtained from patients randomly assigned to control groups.